No abstract

Summary This paper presents a simplified finite element analysis technique, the ‘Press‐Replace’ technique, to model pile penetration problems in geotechnical engineering, particularly, pile jacking. The method is employed in standard finite element analysis software. The method involves a straining and a consequent geometry update phase. First, a cone penetration test in (undrained) clay is modelled and compared with the results of analytical, semi‐analytical and more advanced finite element techniques. The model sensitivity for the step size and mesh is investigated using a hypoplastic constitutive model. An optimum way of modelling based on the numerical performance is shown. Copyright © 2015 John Wiley & Sons, Ltd.

No abstract

The relative density is an important state parameter in the determination of engineering properties of marine sands. Existing industry-acknowledged CPT correlations to relative density have been developed almost exclusively from the results of calibration chamber testing performed at stress levels higher than 50 kPa. A few studies suggest correlations for low stress levels as well, however mainly for unaged normally consolidated soils. Existing formulations are often found to be challenged in the application for dense to very dense aged overconsolidated marine sands. This paper investigates the conditions and response of overconsolidated medium dense to very dense sand (as typically found in the North Sea) at shallow depths (less than 50 kPa) for a penetrating cone. It provides recommendations for a consistent set of parameters linking CPT parameters to the stress conditions, expressed by the apparent overconsolidation ratio and coefficient of earth pressure at rest, and the relative density. The recommendations are based on a comprehensive field testing campaign at a sand site in Cuxhaven, Germany, supported by a suite of laboratory testing and numerical analyses.

The needle penetrometer (NP) is a nondestructive, cheap and simple device which can be used indirectly to obtain the uniaxial compressive strength of extremely weak rocks. It requires little sample preparation and can be used in the field and laboratory and applied in natural and man-made structures where sampling is not allowed. The microstructural damage created by the needle, its shape and size, have been assessed using four rock types (marl, mudstone and tuff from Turkey and calcarenite from the Netherlands) and two types of needle. During needle penetration, very high compressive and shear stresses are developed in advance of the needle and normal to the shaft. In all the tested rocks, densification occurred in a zone some 0.4 and 1 mm ahead of the Maruto and Eijkelkamp needles respectively. The grains are crushed and compacted in a zone which appears like an extension of the needle tip (Eijkelkamp needle). As damage is local, the NP test is said to be non-destructive. The NP results in coarse grained rocks are far more variable than in fine grained rocks. The damage caused by a needle decreases with a decrease in the diameter of the needle and an increase in the slenderness of the needle.

This paper presents a novel technique to describe the installation effects around a jacked pile without simulating the penetration process. With the proposed technique, a jacked pile can be modelled as wished-in-place, and corresponding installation effects, such as stress and density change, can be imposed. In order to obtain a trend for parameters affecting the installation effects, different pile diameters (D), penetration depths (L) and soil densities (Id) are investigated. A mesh sensitivity analysis is performed to assess the possibilities and limitations of transferring the installation effects (stress and density). Model functions are proposed to approximate stress components and void ratio distribution. After finding geometrical functions for each variation by employing non-linear regression analyses, a generalised form based on D, L and Id is proposed. Finally the generalised formulations are validated by a centrifuge test.

No abstract

No abstract